Open cmdevries opened 7 years ago
Use compiler intrinsics to try different SIMD implementations of Hamming distance for both exlusive or and population count
Can also try mixing SIMD Hamming distance on FPU + POPCNT which runs on the ALU to keep all execution units on the CPU busy.
Use compiler intrinsics to try different SIMD implementations of Hamming distance for both exlusive or and population count